Response Predictive Factors to Anagrelide in Essential Thrombocythemia Diagnosed Patients.

We performed a retrospective analysis of our Spanish database of patients with ET in order to assess the role of different response predictive potential factors to anagrelide treatment. 411 ET patients from 54 sites from February 2005 to August 2006 were included in a retrospective chart review. ET was diagnosed according to PVSG criteria (1997). All patients had started treatment with anagrelide before December 2004, either as a first line or as a second line therapy. The population was divided itself in three groups of risk at diagnosis:

• of high Risk: those patients older than 60 years or with previous history of thrombosis (40.6%);

• of low Risk: patients younger than 60 years with thrombocytosis lower than 1.5· 10⁹/L, without cardiovascular risk factors and previous history of thrombosis (35.8%);

• intermediate risk: those patients who did not reunite the criteria of previously mentioned groups (23.6%).

The response to the treatment was defined as complete remission (CR), when a reduction of platelets′ count were equal or less than 400×10⁹/L, or a reduction higher than 50% with respect to the basal numbers; partial remission (PR), when the platelets′ count was between 400× 10⁹/L and 600×10⁹/L with respect to the basal numbers; and no response, those with a lower reduction than PR or increasing their platelets′count. CR was obtained in 219 patients (53.6%; 95%CI = 48.6–58.5) and PR in 113 (27.6%; 95%CI = 23.4–32.2), giving an overall response (OR) rate of 81.2% [95%CI = 77.0–84.9]. The influence of certain factors (such as age, gender, risk group, platelets levels before treatment with anagrelide, previous cytoreductive untreated patients and maximum anagrelide dose), in the objective response was explored. A chi-square test and a multivariate analysis (logistic regression) were performed. A worse response (p = 0.021) was associated with a higher dose (≥ 2.5 mg per day). A trend (p = 0.103) to better response in patients without previous treatment (anagrelide as a first line) was detected. Additionally a significant association (p = 0.02) between the previously treated with hydroxyurea and the presence of evolutive hematological transformations (myelofibrosis and acute myeloid leukemia/myelodysplastic syndrome) was observed. It is difficult to set a cause effect relation in all these findings due to the retrospective design of the analysis. As a conclusion, the response to anagrelide in ET patients was demonstrated to be independent of the age, gender, platelets′ count and level of previous risk.